Unverified50% confidenceFactExact time
In the large-scale automated neuron interpretation project Cammarata conducted at OpenAI in 2023, the tools available at the time could explain only about 2% of neurons.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn 2023, an Anthropic research team published a paper using sparse autoencoder technology to decompose the internal neuron activations of the Claude model into millions of interpretable conceptual features.67% similarUnverifiedMirror neurons were accidentally discovered in the premotor cortex of macaques in the 1990s by the team of Italian neuroscientist Giacomo Rizzolatti, and they activate both when executing an action and when observing another individual performing the same action.61% similarUnverifiedAnthropic published a paper on superposition in 2022, revealing the phenomenon of individual neurons simultaneously encoding multiple concepts.60% similarUnverifiedSome researchers systematically reviewed the 9094 papers of NeurIPS 2026, and the analysis results show that agents, reinforcement learning fine-tuning and small models are the three directions repeatedly mentioned.57% similarUnverifiedAnthropic published the dictionary learning research Towards Monosemanticity in 2023, attempting to decompose polysemantic neurons into monosemantic features via sparse autoencoders.56% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1045658API
curl https://kongchang.com/api/v1/knowledge/claims/1045658MCP
get_claim(id=1045658)